Abstract
Background
Chronic kidney disease (CKD) is a frequent non-communicable complication in people living with HIV (PLHIV), influenced by antiretroviral therapy (ART), comorbidities, and healthcare setting. The Data Collection on Adverse Events of Anti-HIV Drugs (DAD) CKD risk score is widely applied internationally, while the Thai CKD risk score, developed for the general Thai population, has not been validated in HIV cohorts. This study compared the predictive accuracy of the DAD models and the Thai CKD risk score in Thai people living with HIV receiving ART.
Methods
A retrospective cohort analysis was conducted using electronic medical records of adults (≥18 years) living with HIV at Warinchamrap Hospital between January 2020 and May 2024. Patients with pre-existing CKD or incomplete data were excluded. CKD was defined per Kidney Disease: Improving Global Outcomes (KDIGO) criteria. Model calibration was assessed with the Hosmer–Lemeshow test, and discrimination with the C-statistic.
Results
Among 901 PLHIV receiving ART, 104 (11.5%) developed incident CKD during a median follow-up of approximately 4 years. Patients who developed CKD were more likely to be male and have a lower baseline estimated glomerular filtration rate, and higher prevalence of proteinuria and comorbidities. All evaluated models stratified CKD risk; however, the full and short DAD scores showed moderate discrimination (C-statistics ≈0.76) with evidence of risk underestimation and suboptimal calibration. The Thai CKD risk score demonstrated higher discrimination (C-statistic 0.93), sensitivity (88.5%), and better calibration in this cohort.
Conclusions
In this Thai HIV cohort, all evaluated risk models provided CKD risk stratification, but the Thai CKD risk score showed more reliable discrimination and calibration than the DAD models. These findings highlight the importance of population-specific validation when applying CKD risk prediction tools in HIV care.
Get full access to this article
View all access options for this article.